具有增广状态向量的离散时间卡尔曼滤波器

时间:2015-04-26 19:04:12

标签: matlab kalman-filter

我正在尝试使用增强状态向量为状态空间模型实现离散时间卡尔曼滤波器。也就是说,状态方程由下式给出:

x (t)= 0.80 x (t-1)+ 噪音

我的示例中只有潜在的过程,因此 x (t)的维度为1x1。测量方程式由下式给出:

Y (t)= A X (t)+ 噪音

其中资本 Y (t)的维度为3x1。它包含三个观察到的序列, Y (t)= [ y1 (t), y2 (t), y3 (t)的]&#39 ;.观察到的两个系列偶尔会以系统的方式丢失:

    在所有时间点都观察到
  • y1 (t)。
  • 每隔四个时间点观察到
  • y2 (t)。
  • 每隔十二个时间点观察到
  • y3 (t)。

测量矩阵 A 的维数为3x12,资本 X (t)的维数为12x1。这是一个堆积了潜在过程的最后十二个实现的向量:

X (t)= [ x (t), x (t-1),..., x (T-11)]&#39 ;.

资本 X (t)堆叠的原因是 y1 (t)与 x (t)和 X (T-1)。此外, y2 (t)与 x (t), x (t-1), x 有关(t-2)和 x (t-3)。最后, y3 (t)与潜在过程的最后十二个实现相关。

附加的matlab代码模拟来自此状态空间模型的数据,然后通过具有增强状态空间矢量 X (t)的卡尔曼滤波器运行。

我的问题是,过滤的(和预测的)过程与真正的潜在过程大不相同。这也可以从附图中看出。显然,我做错了什么,但我无法看到它是什么?当状态向量没有堆叠时,我的卡尔曼滤波器工作得很好......我希望有人可以提供帮助,谢谢!enter image description here

    %-------------------------------------- %
    % AUGMENTED KALMAN FILTER WITH MISSINGS
    % ------------------------------------- %

    clear; close all; clc;

    % 1) SET PARAMETER VALUES %
    % ----------------------- %

    % Number of observations, number of latent processes and observed
    % processes.
    T = 2000;
    NumberOfLatent = 1;
    NumberOfObs = 3;

    % Measurement Matrix, A.
    A = [-0.40, -0.15, zeros(1,10);
          0.25,  0.25, 0.25, 0.25, zeros(1,8);
         ones(1,12)];

    % State transition matrix, Phi. 
    Phi = zeros(12,12);
    Phi(1,1) = 0.80;
    Phi(2:end,1:end-1) = eye(11);

    % Covariance matrix (for the measurement equation).            
    R = diag([2.25; 1.00; 1.00]);

    % State noise and measurement noise.
    MeasNoise = mvnrnd(zeros(1,NumberOfObs),R,T)';
    StateNoise = randn(NumberOfLatent,T+11);

    % One latent process (X) and three observed processes (Y).
    X = zeros(NumberOfObs,T+11);  
    Y = zeros(NumberOfObs,T); 

    % Set initial state.
    X0 = 0;

    % 2) SIMULATE DATA %
    % ---------------- %
    % Before Y (the observed processes) can be simulated we need to simulate 11
    % realisations of X (the latent process).  
    for t = 1:T+11
       if (t == 1)
            X(:,t) = Phi(1,1)*X0 + StateNoise(:,t);
       else
            X(:,t) = Phi(1,1)*X(1,t-1) + StateNoise(:,t);
       end
       if (t>=12)
            Y(:,t-11) = A*X(1,t:-1:t-11)' + MeasNoise(:,t-11);
       end
    end


    % 3) DELETE DATA SUCH THAT THERE ARE MISSINGS %
    % ------------------------------------------- %
    % First row is observed at all time points. The second rows is observed  at
    % every 4th. time point and the third row is observed at every 12th. time
    % point.
    for j = 1:3
        Y(2,j:4:end) = NaN;
    end
    for j = 1:11
        Y(3,j:12:end) = NaN;
    end


    % 4) RUN THE KALMAN FILTER WITH AUGMENTED STATE VECTOR %
    % ---------------------------------------------------- %

    % Initializing matrices.
    AugmentedStates = 12;
    X_predicted = zeros(AugmentedStates,T);                   % Predicted  states.
    X_filtered = zeros(AugmentedStates,T);            % Updated states.
    P_predicted = zeros(AugmentedStates,AugmentedStates,T);   % Predicted variances.
    P_filtered = zeros(AugmentedStates,AugmentedStates,T);   % Filtered variances.
    KG = zeros(AugmentedStates,NumberOfObs,T);       % Kalman gains.
    Q  = eye(AugmentedStates);                       % State noise for  augmented states.
    Res         = zeros(NumberOfObs,T);                       % Residuals.

    % Initial values:
    X_zero = ones(AugmentedStates,1);
    P_zero = eye(AugmentedStates);

    for t = 1:T

        % Initialisation of the Kalman filter.
        if (t == 1)
            X_predicted(:,t) = Phi * X_zero;                          %  Initial predicted state, dimension (AugmentedStates x 1).
            P_predicted(:,:,t) = Phi * P_zero * Phi'+ Q;              % Initial variance, dimension (AugmentedStates x AugmentedStates).
        else
         % #1 Prediction step.
           X_predicted(:,t) = Phi * X_filtered(:,t-1);                    %   Predicted state, dimension (AugmentedStates x 1).
           P_predicted(:,:,t) = Phi * P_filtered(:,:,t-1) * Phi'+ Q;      % Variance of prediction, dimension (AugmentedStates x AugmentedStates).
        end

        % If there is missings the corresponding entries in A and Y is set  to
        % zero.
        if sum(isnan(Y(:,t)))>0
            temp = find(isnan(Y(:,t)));                             
            Y(temp,t) = 0;                                         
            A(temp,:) = 0;              
        end

        % #3 Calculate the Kalman Gains and save them in the matrix KG.
        K = (P_predicted(:,:,t) * A')/(A * P_predicted(:,:,t) * A' + R);         % Kalman Gains, dimension (AugmentedStates x ObservedProcesses).
        KG(:,:,t) = K;  

        % Residuals.
        Res(:,t) = Y(:,t) - A*X_predicted(:,t);                

        % #3 Update step.
        X_filtered(:,t) =  X_predicted(:,t) + K * Res(:,t);                           % Updated state (AugmentedStates x 1).
        P_filtered(:,:,t) = (eye(AugmentedStates) - K * A) * P_predicted(:,:,t);    % Updated variance, dimension (AugmentedStatesstates x AugmentedStates).

         % Measurement Matrix, A, is recreated.
         A = [-0.40, -0.15, zeros(1,10);
               0.25,  0.25, 0.25, 0.25, zeros(1,8);
               ones(1,12)];

     end

1 个答案:

答案 0 :(得分:0)

啊哈。我意识到了这个bug是什么。 matlab代码中的状态转换矩阵(Phi)不正确。也就是说,在当前的matlab代码中它是:

% State transition matrix, Phi. 
Phi = zeros(12,12);
Phi(1,1) = 0.80;
Phi(2:end,1) = 1

它应该看起来像这样:

% State transition matrix, Phi. 
Phi = zeros(12,12);
Phi(1,1) = 0.80;
Phi(2:end,1:end-1) = eye(11)

如果没有这个调整,过滤器中的状态方程就没有任何意义!我在代码中实现了这一点。为了表明它确实有效,情节现在看起来像这样:

enter image description here

......或者至少:过滤器的性能得到了提升!